Publication: Sistema de control de movimientos de brazos robóticos mediante el análisis de la actividad cerebral
Loading...
Identifiers
Publication date
2022-07
Defense date
2022-07-15
Authors
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
En la presente memoria se describe el proyecto Sistema de control de movimientos de brazos robóticos mediante el análisis de la actividad cerebral cuyo objetivo es el de probar la hipótesis de si es suficiente el uso de las ondas cerebrales para determinar el movimiento. Para esto se utilizará una diadema que obtiene el EEG del paciente, una aplicación de lectura y un código de análisis. En ella se presentarán conceptos relacionados con las ondas cerebrales y el análisis de datos. Tras hacer un repaso de la historia del encefalograma y un resumen de las características de las ondas cerebrales, se introducen los diferentes modelos de regresores para el análisis de los datos: KNN, Random Forest, Ridge Regressor y Kernel Ridge Regressor.
En la parte práctica del trabajo se analizan los modelos introduciendo conjuntos de datos con dos, tres y cuatro etiquetas diferentes. Se estudian los resultados y la viabilidad del proyecto en función de la exactitud de los modelos y de los costes asociados. Finalmente, en la última parte se presenta una posible aplicación futura que facilitaría mucho el día a día de pacientes con algún tipo de discapacidad motora.
This study describes the Robotic Arm Movement Control System project by analyzing brain activity, which objective is to test the following hyphotesis: are brain waves enough to determine movement?. For this, a headband that obtains the patient's EEG, a reading application and an analysis code are used. This memory is going to present concepts related to brain waves and data analysis. After reviewing the history of the encephalogram and a summary of the characteristics of brain waves, the different regressor models for data analysis are introduced: KNN, Random Forest, Ridge Regressor and Kernel Ridge Regressor. In the practical part of the work, the models are analyzed by introducing datasets with two, three and four different labels. The results and feasibility of the project are studied based on the accuracy of the models and the associated costs. Finally, in the last part, a possible future application is presented that would greatly facilitate the daily life of patients with some type of motor disability.
This study describes the Robotic Arm Movement Control System project by analyzing brain activity, which objective is to test the following hyphotesis: are brain waves enough to determine movement?. For this, a headband that obtains the patient's EEG, a reading application and an analysis code are used. This memory is going to present concepts related to brain waves and data analysis. After reviewing the history of the encephalogram and a summary of the characteristics of brain waves, the different regressor models for data analysis are introduced: KNN, Random Forest, Ridge Regressor and Kernel Ridge Regressor. In the practical part of the work, the models are analyzed by introducing datasets with two, three and four different labels. The results and feasibility of the project are studied based on the accuracy of the models and the associated costs. Finally, in the last part, a possible future application is presented that would greatly facilitate the daily life of patients with some type of motor disability.
Description
Keywords
Inteligencia artificial, Brazos robóticos, Neurosky, Kernel Ridge Regressor, Random Forest, KNN, Artificial intelligence